repo stringclasses 1
value | github_id int64 3.77B 4.38B | github_node_id stringlengths 18 24 | number int64 75 77.8k | html_url stringlengths 46 49 | api_url stringlengths 56 59 | title stringlengths 1 746 | body stringlengths 1 55.7k ⌀ | state stringclasses 2
values | state_reason stringclasses 4
values | locked bool 2
classes | comments_count int64 0 128 | labels listlengths 0 5 | assignees listlengths 0 2 | created_at stringdate 2026-01-01 16:55:34 2026-05-05 11:59:03 | updated_at stringdate 2026-03-24 00:13:29 2026-05-05 11:59:15 | closed_at stringlengths 20 20 ⌀ | author_association stringclasses 3
values | milestone_title stringclasses 0
values | snapshot_id stringclasses 76
values | extracted_at stringdate 2026-03-24 23:36:49 2026-05-05 12:00:33 | author_login stringlengths 3 38 | author_id int64 346 282M | author_node_id stringlengths 12 20 | author_type stringclasses 2
values | author_site_admin bool 2
classes |
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
openclaw/openclaw | 4,364,199,632 | I_kwDOQb6kR88AAAABBCBm0A | 75,618 | https://github.com/openclaw/openclaw/issues/75618 | https://api.github.com/repos/openclaw/openclaw/issues/75618 | [2026.4.23] Feishu multi-account runtime drops top-level allowlist fields for accounts.main | # Feishu multi-account runtime drops top-level allowlist fields for `accounts.main` in 2026.4.23
## Environment
- OpenClaw version: `2026.4.23 (a979721)`
- Channel: `feishu`
- Config shape: `channels.feishu.accounts.main` is present
- Runtime mode: reproduced in a real gateway runtime while handling Feishu group mess... | closed | completed | true | 1 | [] | [] | 2026-05-01T11:42:48Z | 2026-05-04T04:36:39Z | 2026-05-01T11:46:35Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | crazyn2 | 45,783,572 | MDQ6VXNlcjQ1NzgzNTcy | User | false |
openclaw/openclaw | 4,364,176,969 | I_kwDOQb6kR88AAAABBCAOSQ | 75,616 | https://github.com/openclaw/openclaw/issues/75616 | https://api.github.com/repos/openclaw/openclaw/issues/75616 | Per-request startup regression in 4.29: ~25–50s overhead before model output (core-plugin-tools 11s, auth 4–12s, model-resolution 8s) | ## Summary
`v2026.4.29` introduced a per-request startup regression that adds **25–50 seconds of overhead before the model starts streaming**. It's reproducible on every embedded-run, not a one-off cold-start cost. Rolling back to `v2026.4.23` restores normal latency.
## Symptoms after upgrading 4.23 → 4.29
User sen... | closed | completed | true | 2 | [] | [] | 2026-05-01T11:36:36Z | 2026-05-04T04:36:38Z | 2026-05-01T11:47:48Z | CONTRIBUTOR | null | 20260504T060031Z | 2026-05-04T06:00:31Z | Udjin79 | 68,610,242 | MDQ6VXNlcjY4NjEwMjQy | User | false |
openclaw/openclaw | 4,362,222,686 | I_kwDOQb6kR88AAAABBAI8Xg | 75,330 | https://github.com/openclaw/openclaw/issues/75330 | https://api.github.com/repos/openclaw/openclaw/issues/75330 | Gateway event loop blocked during agent prep: max delay 32s, utilization 100%, API requests hang | ## Version
OpenClaw 2026.4.29 (a448042)
## Environment
- OS: CachyOS (Arch Linux), Kernel 6.19.12-1
- Node: v25.9.0
- Install: npm global
- Hardware: Dell Latitude 5300
## Symptoms
Gateway API requests become extremely slow or hang during agent preparation. The event loop shows massive blocking spikes. The `allgemein... | closed | completed | true | 9 | [] | [] | 2026-05-01T01:05:01Z | 2026-05-04T04:36:38Z | 2026-05-01T11:47:52Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | erforschtbot-cmyk | 272,903,176 | U_kgDOEEQsCA | User | false |
openclaw/openclaw | 4,362,863,229 | I_kwDOQb6kR88AAAABBAwCfQ | 75,428 | https://github.com/openclaw/openclaw/issues/75428 | https://api.github.com/repos/openclaw/openclaw/issues/75428 | [Bug]: answer question too slow,when update to 4.29 | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
after update to 4.29 ,weixin channel answer question become too slow ,and i flow logs ,get below info:[trace:embedded-run] prep stages: runId=b857a808-a622-41b8-b623-9d87f1edcaee sessionId=269f8677-bd7a-4a84-83d4-6f2f3536eab... | closed | completed | true | 3 | [
"bug",
"regression"
] | [] | 2026-05-01T04:48:56Z | 2026-05-04T04:36:36Z | 2026-05-01T11:47:57Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | steven0081 | 42,331,427 | MDQ6VXNlcjQyMzMxNDI3 | User | false |
openclaw/openclaw | 4,363,998,916 | I_kwDOQb6kR88AAAABBB1WxA | 75,598 | https://github.com/openclaw/openclaw/issues/75598 | https://api.github.com/repos/openclaw/openclaw/issues/75598 | [Bug]: “EventLoopDelayMaxMs spikes 6–24s + stuck sessions on low-power N355 Proxmox after 4.23” | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
On OpenClaw 2026.4.23 the gateway is fast and responsive. After upgrading to any version 4.25+, periodic multi-second eventLoopDelayMaxMs spikes (6–24+ seconds) and stuck-session warnings appear even in clean sessions.
### ... | closed | completed | true | 10 | [
"bug",
"regression"
] | [] | 2026-05-01T10:43:02Z | 2026-05-04T04:33:34Z | 2026-05-01T11:48:00Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | trojy13 | 93,163,196 | U_kgDOBY2OvA | User | false |
openclaw/openclaw | 4,360,549,388 | I_kwDOQb6kR88AAAABA-i0DA | 75,214 | https://github.com/openclaw/openclaw/issues/75214 | https://api.github.com/repos/openclaw/openclaw/issues/75214 | [Bug]: Gateway event-loop stalls during | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
On a fresh OpenClaw `2026.4.27` install, the Gateway repeatedly emits diagnostic liveness warnings showing multi-second Node event-loop stalls.
The stalls correlate with Gateway RPC paths including:
- `chat.history`
- `log... | closed | completed | true | 2 | [
"bug",
"regression"
] | [] | 2026-04-30T18:53:51Z | 2026-05-04T04:36:32Z | 2026-05-01T11:49:07Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | groff-lab | 280,176,582 | U_kgDOELMnxg | User | false |
openclaw/openclaw | 4,361,658,522 | I_kwDOQb6kR88AAAABA_mgmg | 75,290 | https://github.com/openclaw/openclaw/issues/75290 | https://api.github.com/repos/openclaw/openclaw/issues/75290 | core-plugin-tools prep stage takes ~11s on every embedded run (cached registry, per-run factory cost) | ## Summary
In `runEmbeddedAttempt`, the prep stage `core-plugin-tools` consistently takes **~10–12 s on every embedded run**, dominating startup latency. The cost is paid even when the plugin registry is fully cached (no manifest scan), so it's not a discovery/IO cost — it's coming from per-run plugin tool factory inv... | closed | completed | true | 14 | [] | [] | 2026-04-30T22:33:13Z | 2026-05-05T04:25:11Z | 2026-05-02T18:30:28Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | thanos-openclaw | 261,652,258 | U_kgDOD5h_Ig | User | false |
openclaw/openclaw | 4,364,260,496 | I_kwDOQb6kR88AAAABBCFUkA | 75,624 | https://github.com/openclaw/openclaw/issues/75624 | https://api.github.com/repos/openclaw/openclaw/issues/75624 | Misleading "sqlite-vec unavailable" warning when sqlite-vec actually loads successfully | ## Summary
`chunks_vec not updated — sqlite-vec unavailable. Vector recall degraded.` is logged even when the sqlite-vec extension successfully loads. The warning fires whenever `vectorReady === false`, which can be caused by reasons unrelated to sqlite-vec (e.g., missing embedding provider). This makes diagnosing the ... | open | null | false | 1 | [] | [] | 2026-05-01T11:59:21Z | 2026-05-03T05:00:49Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | Noah3521 | 115,360,587 | U_kgDOBuBDSw | User | false |
openclaw/openclaw | 4,363,997,582 | I_kwDOQb6kR88AAAABBB1Rjg | 75,596 | https://github.com/openclaw/openclaw/issues/75596 | https://api.github.com/repos/openclaw/openclaw/issues/75596 | [Feature Request] WeChat plugin: support streaming message sending (real-time output) | ## Problem
When using the WeChat plugin, the entire AI response is sent as one message after the model finishes generating. Even with `blockStreaming: false` in config, users dont see the response appear piece by piece — they only see it when the full response is ready.
This means users have to wait a long time withou... | closed | not_planned | true | 1 | [] | [] | 2026-05-01T10:42:36Z | 2026-05-04T04:36:30Z | 2026-05-01T12:06:54Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | ryswork1993 | 44,529,247 | MDQ6VXNlcjQ0NTI5MjQ3 | User | false |
openclaw/openclaw | 4,364,230,492 | I_kwDOQb6kR88AAAABBCDfXA | 75,621 | https://github.com/openclaw/openclaw/issues/75621 | https://api.github.com/repos/openclaw/openclaw/issues/75621 | Gateway lazy-spawns duplicate stdio MCP children with identical ppid+config (memory + CPU leak) | ## Summary
On 2026.4.29, the OpenClaw gateway sometimes lazy-spawns **two** stdio MCP child processes for the same configured server, with identical `ppid` and identical `--config` argument. Both children stay alive indefinitely, doubling memory cost and CPU during the index-rebuild / handshake phase that runs at chil... | open | null | false | 3 | [] | [] | 2026-05-01T11:51:48Z | 2026-05-04T21:36:42Z | null | NONE | null | 20260505T000034Z | 2026-05-05T00:00:34Z | zackchiutw | 107,488,700 | U_kgDOBmglvA | User | false |
openclaw/openclaw | 4,363,142,335 | I_kwDOQb6kR88AAAABBBBEvw | 75,468 | https://github.com/openclaw/openclaw/issues/75468 | https://api.github.com/repos/openclaw/openclaw/issues/75468 | [Feature]: Agent safeguards - content hash dedup for file writes + general-purpose max iteration limit | ## Summary
An autonomous agent session produced **382 identical copies** of the same file (all 2938 bytes, same content) over ~37 minutes — writing once every ~6 seconds — before being manually reset. The agent was stuck in a "check → hallucinate improvement needed → rewrite same content with versioned filename → repe... | open | null | false | 1 | [] | [] | 2026-05-01T06:20:08Z | 2026-05-03T10:41:59Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | aiguo-max | 256,335,049 | U_kgDOD0dcyQ | User | false |
openclaw/openclaw | 4,364,348,126 | I_kwDOQb6kR88AAAABBCKq3g | 75,631 | https://github.com/openclaw/openclaw/issues/75631 | https://api.github.com/repos/openclaw/openclaw/issues/75631 | macOS app/dashboard config writes can drop gateway.auth and restart Gateway | ## Summary
The macOS app / dashboard configuration path can rewrite `~/.openclaw/openclaw.json` in a way that drops existing gateway auth configuration and triggers a Gateway restart. Active Control UI / dashboard WebSocket clients are then disconnected with `1012 service restart`.
This looks related to the existing ... | closed | completed | true | 1 | [] | [] | 2026-05-01T12:24:34Z | 2026-05-05T04:28:54Z | 2026-05-02T04:45:57Z | CONTRIBUTOR | null | 20260505T060034Z | 2026-05-05T06:00:34Z | Fuma2013 | 6,313,918 | MDQ6VXNlcjYzMTM5MTg= | User | false |
openclaw/openclaw | 4,364,352,770 | I_kwDOQb6kR88AAAABBCK9Ag | 75,633 | https://github.com/openclaw/openclaw/issues/75633 | https://api.github.com/repos/openclaw/openclaw/issues/75633 | [Bug]: Anthropic-messages adapter crashes with "Cannot read properties of undefined (reading 'type')" on MiniMax Anthropic-compatible endpoint` | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
## Environment
- OpenClaw: `2026.4.21 (f788c88)`
- Node: `v24.14.1`
- OS: Windows 11
- Provider config: `api: "anthropic-messages"`, `baseUrl: "https://api.minimax.io/anthropic"`, `apiKey` from env (sk-cp- Coding Plan key)
-... | closed | completed | true | 2 | [
"bug"
] | [] | 2026-05-01T12:26:02Z | 2026-05-05T04:27:53Z | 2026-05-02T09:36:48Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | cmancil | 132,243,972 | U_kgDOB-HiBA | User | false |
openclaw/openclaw | 4,364,350,998 | I_kwDOQb6kR88AAAABBCK2Fg | 75,632 | https://github.com/openclaw/openclaw/issues/75632 | https://api.github.com/repos/openclaw/openclaw/issues/75632 | [Bug]: TUI first-connect / chat.history / models.list stalls on 2026.4.27-2026.4.29, rollback to 2026.4.23 massively improves startup | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
TUI startup and websocket initialization are much slower and less reliable on OpenClaw 2026.4.27 and 2026.4.29 than on 2026.4.23, with isolated --dev tests showing handshake timeouts plus slow chat.history/mode... | closed | completed | true | 1 | [
"bug",
"bug:behavior"
] | [] | 2026-05-01T12:25:28Z | 2026-05-04T04:36:30Z | 2026-05-01T12:30:11Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | oclavvbox-cyber | 263,183,708 | U_kgDOD6_dXA | User | false |
openclaw/openclaw | 4,364,353,892 | I_kwDOQb6kR88AAAABBCLBZA | 75,634 | https://github.com/openclaw/openclaw/issues/75634 | https://api.github.com/repos/openclaw/openclaw/issues/75634 | [Bug]: Sharp prebuilt binary fails on Common KVM / SSE2-only CPU (N355) — recurring "Failed to optimize image" after every update | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
On virtualized Common KVM processor (SSE2 only, no v2 microarchitecture), Sharp’s prebuilt linux-x64 binaries fail to load. Image optimization throws "Failed to optimize image". Custom try/catch + libvips fallback works but ... | open | null | false | 1 | [
"bug",
"regression"
] | [] | 2026-05-01T12:26:22Z | 2026-05-03T09:13:44Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | trojy13 | 93,163,196 | U_kgDOBY2OvA | User | false |
openclaw/openclaw | 4,364,371,953 | I_kwDOQb6kR88AAAABBCMH8Q | 75,636 | https://github.com/openclaw/openclaw/issues/75636 | https://api.github.com/repos/openclaw/openclaw/issues/75636 | [Bug]: 4.29 still crashes plugin-runtime-deps with `Invalid Version: ''` on Node v22 / npm v10 (different code path from #74949) | ## Summary
On `openclaw@2026.4.29` (final, not beta), a fresh install on **Node v22.22.0 / npm v10.9.4 (macOS)** still crashes the bundled plugin-runtime-deps installer with `TypeError: Invalid Version: ''`. The fix from #74949 is present in the shipped tarball (the generated `package.json` correctly populates `depend... | closed | completed | true | 2 | [] | [] | 2026-05-01T12:31:43Z | 2026-05-04T04:35:36Z | 2026-05-01T20:42:46Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | Sleepyarno | 112,829,252 | U_kgDOBrmjRA | User | false |
openclaw/openclaw | 4,364,285,745 | I_kwDOQb6kR88AAAABBCG3MQ | 75,626 | https://github.com/openclaw/openclaw/issues/75626 | https://api.github.com/repos/openclaw/openclaw/issues/75626 | feat(oap-sidecar): ship /usr/local/bin/oap-healthcheck binary for portable container healthchecks | ## Summary
Request that the `ghcr.io/openclaw/oap-sidecar` image ship a small `/usr/local/bin/oap-healthcheck` binary so consumers running the sidecar in container orchestrators (ECS, Kubernetes, Nomad) can configure a healthcheck that does not depend on the host shell or curl.
## Why this matters
Today, consumers w... | closed | not_planned | true | 1 | [] | [] | 2026-05-01T12:06:54Z | 2026-05-04T04:36:27Z | 2026-05-01T12:36:08Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | ender-wiggin-ai | 266,379,588 | U_kgDOD-ChRA | User | false |
openclaw/openclaw | 4,364,416,458 | I_kwDOQb6kR88AAAABBCO1yg | 75,639 | https://github.com/openclaw/openclaw/issues/75639 | https://api.github.com/repos/openclaw/openclaw/issues/75639 | Follow-up to #72875: local memory canary on 2026.4.27 triggers broad channel runtime-deps installs and gateway probe timeouts | Follow-up to #72875.
We originally hit the `memorySearch.provider = "local"` path on 2026.4.23, where the memory runtime failed with `Unknown memory embedding provider: local` even though the capability embedding path could see and use the local provider.
I moved this same VPS to 2026.4.27 because the recent release ... | closed | completed | true | 1 | [] | [] | 2026-05-01T12:44:50Z | 2026-05-04T04:36:25Z | 2026-05-01T12:49:47Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | Amirofcodes | 138,374,972 | U_kgDOCD9vPA | User | false |
openclaw/openclaw | 4,364,431,044 | I_kwDOQb6kR88AAAABBCPuxA | 75,641 | https://github.com/openclaw/openclaw/issues/75641 | https://api.github.com/repos/openclaw/openclaw/issues/75641 | Codex harness: project live tool calls as Control UI / WebChat tool cards | ## Summary
When using the Codex harness (`codex/*` model refs) from Control UI / WebChat, tool execution appears to run correctly, but the UI does not show Codex-style live tool-call cards or tool progress. The user only sees assistant text updates and the final response, which makes long tool-using runs feel opaque c... | open | null | false | 1 | [] | [] | 2026-05-01T12:48:49Z | 2026-05-03T07:45:47Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | ygc3817922006-sketch | 266,983,010 | U_kgDOD-nWYg | User | false |
openclaw/openclaw | 4,363,422,584 | I_kwDOQb6kR88AAAABBBSLeA | 75,512 | https://github.com/openclaw/openclaw/issues/75512 | https://api.github.com/repos/openclaw/openclaw/issues/75512 | [Bug]: Chat-turn latency 30-60s+ on every turn since v2026.4.23: per-turn plugin runtime re-evaluation (cross-platform BLOCKER) | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
After upgrading from v2026.4.23 to v2026.4.26, every chat turn's embedded-runner prep phase takes ~50+ seconds (LLM call ~3s; total ~76s) on ARM hardware. Befor upgrade was like 15s.
### Steps to reproduce
1. On an ARM SBC... | closed | completed | true | 11 | [
"bug",
"regression"
] | [] | 2026-05-01T07:45:14Z | 2026-05-04T04:33:39Z | 2026-05-01T11:47:50Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | sergeyksv | 1,007,040 | MDQ6VXNlcjEwMDcwNDA= | User | false |
openclaw/openclaw | 4,363,971,448 | I_kwDOQb6kR88AAAABBBzreA | 75,591 | https://github.com/openclaw/openclaw/issues/75591 | https://api.github.com/repos/openclaw/openclaw/issues/75591 | [Bug]: Each plugin extension manifest read 100+ times per request → 60s prep stages | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
Setup: OpenClaw 2026.4.29 npm global, Hetzner CX33 (4 vCPU, 8GB), 1 channel (telegram), 0 hooks, idle daemon.
Symptom: Every embedded-run takes 90-180s before first token. Single-word "hi" message → 1m53s cons... | closed | duplicate | true | 8 | [
"bug",
"bug:behavior"
] | [] | 2026-05-01T10:34:29Z | 2026-05-04T04:35:12Z | 2026-05-01T22:26:20Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | renatmannanov | 140,875,658 | U_kgDOCGWXig | User | false |
openclaw/openclaw | 4,364,459,490 | I_kwDOQb6kR88AAAABBCRd4g | 75,643 | https://github.com/openclaw/openclaw/issues/75643 | https://api.github.com/repos/openclaw/openclaw/issues/75643 | Webchat: internal runtime tokens (NO_REPLY, cron status lines) flicker visibly before sanitization | # Webchat: internal runtime tokens (NO_REPLY, cron status lines) flicker visibly before sanitization
## Summary
In webchat (control-ui) sessions, internal runtime tokens stream into the rendered conversation as they arrive, then get stripped at end-of-message by the normalizer. Result: brief but visible flicker of in... | closed | not_planned | true | 1 | [] | [] | 2026-05-01T12:55:54Z | 2026-05-04T04:36:24Z | 2026-05-01T13:01:51Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | typosiron-png | 273,796,731 | U_kgDOEFHOew | User | false |
openclaw/openclaw | 4,363,188,387 | I_kwDOQb6kR88AAAABBBD4ow | 75,474 | https://github.com/openclaw/openclaw/issues/75474 | https://api.github.com/repos/openclaw/openclaw/issues/75474 | [Bug]: timeout | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
OpenClaw fails to connect to llm
### Steps to reproduce
Title:
OpenClaw 2026.4.29 embedded agent times out on openai/gpt-5.2 and gpt-5.4 while raw OpenAI Responses/SSE works
Environment:
Ubuntu 24.04 EC2
OpenClaw 2026.4.2... | closed | completed | true | 3 | [
"bug",
"regression"
] | [] | 2026-05-01T06:33:23Z | 2026-05-04T04:33:35Z | 2026-05-01T12:06:49Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | ri222AI | 180,990,982 | U_kgDOCsm0Bg | User | false |
openclaw/openclaw | 4,360,736,459 | I_kwDOQb6kR88AAAABA-uOyw | 75,235 | https://github.com/openclaw/openclaw/issues/75235 | https://api.github.com/repos/openclaw/openclaw/issues/75235 | Session corruption: leading-assistant transcript causes infinite "messages: at least one message is required" loop | ## Summary
A session can enter a permanently broken state where every subsequent turn fails with a 400 from the Anthropic API:
```
400 invalid_request_error: messages: at least one message is required
```
The runtime detects this as `failoverReason: "format"` / `providerRuntimeFailureKind: "schema"` and surfaces the... | closed | completed | true | 1 | [] | [] | 2026-04-30T19:29:09Z | 2026-05-04T04:36:23Z | 2026-05-01T13:14:19Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | EVEHetzner | 260,284,917 | U_kgDOD4Oh9Q | User | false |
openclaw/openclaw | 4,364,518,026 | I_kwDOQb6kR88AAAABBCVCig | 75,648 | https://github.com/openclaw/openclaw/issues/75648 | https://api.github.com/repos/openclaw/openclaw/issues/75648 | [Enhancement] Embedded-run upstream timeout is hard-coded at ~60s; provider 429/Retry-After is erased | ## Summary
In a self-hosted OpenClaw deployment we've observed that the gateway's
**embedded-run path enforces a hard ~60-second timeout on the upstream
provider call**, with no configuration knob exposed via `openclaw.json`,
helm chart values, or pod environment variables.
When upstream providers (Anthropic / OpenAI... | open | null | false | 1 | [] | [] | 2026-05-01T13:09:58Z | 2026-05-03T06:08:20Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | lesdekock | 11,728,360 | MDQ6VXNlcjExNzI4MzYw | User | false |
openclaw/openclaw | 4,360,047,482 | I_kwDOQb6kR88AAAABA-ELeg | 75,182 | https://github.com/openclaw/openclaw/issues/75182 | https://api.github.com/repos/openclaw/openclaw/issues/75182 | [Bug]: MEDIA directive parser can include trailing serialized JSON in local file path | ## Bug type
Behavior bug (incorrect output/state without crash)
## Beta release blocker
No
## Summary
A MEDIA directive containing a local image path followed by serialized tool-result JSON can be parsed as one polluted file path, causing outbound media delivery to fail with ENOENT.
## Steps to reproduce
1. Gene... | closed | completed | true | 2 | [] | [] | 2026-04-30T17:30:13Z | 2026-05-04T04:34:12Z | 2026-05-02T03:42:28Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | TnzGit | 24,201,878 | MDQ6VXNlcjI0MjAxODc4 | User | false |
openclaw/openclaw | 4,360,136,055 | I_kwDOQb6kR88AAAABA-Jldw | 75,196 | https://github.com/openclaw/openclaw/issues/75196 | https://api.github.com/repos/openclaw/openclaw/issues/75196 | Subagent completion event sometimes drops result payload and/or token stats (3 repros, intermittent) | ## Summary
Subagent runs (`runtime: "subagent"`, `agentId: "worker"`) that complete normally — committing to git and pushing to origin — sometimes deliver a **completion event with an empty/zero result payload** instead of the structured final-status block the worker actually produced. The work itself is fine; only th... | open | null | false | 2 | [] | [] | 2026-04-30T17:45:54Z | 2026-05-03T05:00:38Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | DavidLeVineASC | 128,732,307 | U_kgDOB6xMkw | User | false |
openclaw/openclaw | 4,360,586,723 | I_kwDOQb6kR88AAAABA-lF4w | 75,220 | https://github.com/openclaw/openclaw/issues/75220 | https://api.github.com/repos/openclaw/openclaw/issues/75220 | [Bug]: gateway install --force generates PATH entries that doctor flags as non-minimal | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
openclaw gateway install --force generates a gateway service PATH that openclaw doctor immediately warns is non-minimal, so the repair path recreates the warned-about state.
### Steps to reproduce
1. Run open... | closed | completed | true | 3 | [
"bug"
] | [] | 2026-04-30T19:00:32Z | 2026-05-05T04:27:38Z | 2026-05-02T10:50:52Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | renaudcerrato | 1,930,949 | MDQ6VXNlcjE5MzA5NDk= | User | false |
openclaw/openclaw | 4,360,722,639 | I_kwDOQb6kR88AAAABA-tYzw | 75,233 | https://github.com/openclaw/openclaw/issues/75233 | https://api.github.com/repos/openclaw/openclaw/issues/75233 | [Bug]: Gateway LaunchAgent plist captures interactive-shell PATH including Node version managers (Volta/asdf/fnm/pnpm), causing non-deterministic Node resolution and update-flow timeouts | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
On macOS, the LaunchAgent plist for `ai.openclaw.gateway` captures the interactive-shell PATH including Node version managers (Volta, asdf, fnm, pnpm). `openclaw gateway status --deep` flags this as a "Service ... | closed | completed | true | 2 | [
"bug",
"bug:behavior"
] | [] | 2026-04-30T19:26:27Z | 2026-05-05T04:27:35Z | 2026-05-02T11:03:26Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | nphyde2 | 269,757,124 | U_kgDOEBQqxA | User | false |
openclaw/openclaw | 4,360,753,647 | I_kwDOQb6kR88AAAABA-vR7w | 75,236 | https://github.com/openclaw/openclaw/issues/75236 | https://api.github.com/repos/openclaw/openclaw/issues/75236 | [Feature]: feat(gateway): apply cursor pagination to agents.files.list and sessions.list | ### Summary
artifacts.list gained cursor pagination in #75208 via src/gateway/util/cursor.ts.
The same pattern should be applied to agents.files.list and sessions list RPCs
to keep the API surface consistent.
Refs #75208.
### Problem to solve
agents.files.list and sessions list RPCs return unbounded result sets.
L... | open | null | false | 1 | [
"enhancement"
] | [] | 2026-04-30T19:31:50Z | 2026-05-03T04:44:33Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | Angfr95 | 175,496,729 | U_kgDOCnXeGQ | User | false |
openclaw/openclaw | 4,360,810,896 | I_kwDOQb6kR88AAAABA-yxkA | 75,239 | https://github.com/openclaw/openclaw/issues/75239 | https://api.github.com/repos/openclaw/openclaw/issues/75239 | WebChat assistant replies still duplicate after #66875 fix (2026.4.27) | ## Bug type: Regression (worked before, now fails)
**Issue #66875 fix is incomplete** — WebChat ControlUI still duplicates every assistant reply on 2026.4.27.
## Environment
- OpenClaw: 2026.4.27 (cbc2ba0)
- Channel: WebChat (ControlUI)
- Install: npm global
- OS: macOS (arm64)
- Browser: Chrome
## Steps to reprodu... | open | null | false | 1 | [] | [] | 2026-04-30T19:43:13Z | 2026-05-03T01:23:29Z | null | NONE | null | 20260503T030030Z | 2026-05-03T03:00:30Z | yanguangzhe-collab | 260,921,814 | U_kgDOD41Z1g | User | false |
openclaw/openclaw | 4,361,187,859 | I_kwDOQb6kR88AAAABA_JyEw | 75,261 | https://github.com/openclaw/openclaw/issues/75261 | https://api.github.com/repos/openclaw/openclaw/issues/75261 | [Bug]: cron with sessionTarget=current still lowercases Matrix roomId in 2026.4.27 (regression of #71798 via MCP loopback path) | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
The fix from #71798 (commit `e309fd485e`, released in 2026.4.25) does not cover the case where `mcp__openclaw__cron action=add` with `sessionTarget: "current"` (and no explicit `delivery`) is invoked from a Cla... | open | null | false | 1 | [] | [] | 2026-04-30T20:51:20Z | 2026-05-03T04:44:35Z | null | CONTRIBUTOR | null | 20260503T060030Z | 2026-05-03T06:00:30Z | sk7n4k3d | 82,924,700 | MDQ6VXNlcjgyOTI0NzAw | User | false |
openclaw/openclaw | 4,361,630,008 | I_kwDOQb6kR88AAAABA_kxOA | 75,287 | https://github.com/openclaw/openclaw/issues/75287 | https://api.github.com/repos/openclaw/openclaw/issues/75287 | Unknown CLI commands load plugins and leave hot child process while gateway saturates | ## Summary
A live `main` source checkout is showing two related CPU/event-loop issues:
1. Running a non-existent command such as `openclaw foo` does not fail fast. It loads provider plugins, loads the runtime plugin set, prints an unavailable-command error, and leaves a hot `openclaw` child process alive at ~90% CPU.... | closed | completed | false | 4 | [] | [] | 2026-04-30T22:26:12Z | 2026-05-03T19:06:51Z | 2026-05-03T19:06:51Z | NONE | null | 20260503T210034Z | 2026-05-03T21:00:34Z | bcdonadio | 1,458,878 | MDQ6VXNlcjE0NTg4Nzg= | User | false |
openclaw/openclaw | 4,361,998,798 | I_kwDOQb6kR88AAAABA_7Rzg | 75,312 | https://github.com/openclaw/openclaw/issues/75312 | https://api.github.com/repos/openclaw/openclaw/issues/75312 | Bug: wiki_search throws 'sharedMemoryManager.search is not a function' when search.backend=shared and corpus includes memory/all | ## Bug Report
### Summary
`wiki_search` throws `sharedMemoryManager.search is not a function` when the memory-wiki config has `search.backend = "shared"` and the query triggers shared memory search (corpus="all" or corpus="memory"). This causes wiki_search to return "No results" even when relevant wiki pages exist.
#... | open | null | false | 1 | [] | [] | 2026-04-30T23:59:40Z | 2026-05-03T07:45:44Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | xuanmingguo | 258,405,939 | U_kgDOD2b2Mw | User | false |
openclaw/openclaw | 4,363,079,427 | I_kwDOQb6kR88AAAABBA9PAw | 75,454 | https://github.com/openclaw/openclaw/issues/75454 | https://api.github.com/repos/openclaw/openclaw/issues/75454 | 1M context window not applied for github-copilot/claude-opus-4.x-1m models (capped at 272k) | ## Summary
1M-context Claude Opus models routed via the `github-copilot/` provider (e.g. `github-copilot/claude-opus-4.6-1m`) are reported with a 272k context window in status output, even when the model is configured with `contextWindow: 1000000` in `openclaw.json`. The 1M auto-promotion logic only whitelists `anthrop... | open | reopened | false | 2 | [] | [] | 2026-05-01T05:56:57Z | 2026-05-03T06:08:18Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | saju01 | 30,193,438 | MDQ6VXNlcjMwMTkzNDM4 | User | false |
openclaw/openclaw | 4,363,140,030 | I_kwDOQb6kR88AAAABBBA7vg | 75,467 | https://github.com/openclaw/openclaw/issues/75467 | https://api.github.com/repos/openclaw/openclaw/issues/75467 | unsupportedToolSchemaKeywords not applied for OpenAI-completions providers (Fireworks) | ## Problem
`getCompat(model)` in `openai-transport-stream-aPa0aR5w.js` builds a new compat object from `model.compat` but only copies specific known fields — it does NOT include `unsupportedToolSchemaKeywords`. So even when a model's catalog entry has `compat.unsupportedToolSchemaKeywords: ["not"]`, the keyword is str... | open | null | false | 1 | [] | [] | 2026-05-01T06:19:22Z | 2026-05-03T01:23:34Z | null | NONE | null | 20260503T030030Z | 2026-05-03T03:00:30Z | bobrenze-bot | 258,452,996 | U_kgDOD2euBA | User | false |
openclaw/openclaw | 4,363,198,122 | I_kwDOQb6kR88AAAABBBEeqg | 75,477 | https://github.com/openclaw/openclaw/issues/75477 | https://api.github.com/repos/openclaw/openclaw/issues/75477 | Telegram CLI message send can hang instead of delegating to gateway | ## Summary
`openclaw message send --channel telegram ...` can hang or fail to return even when the gateway is healthy and direct `openclaw gateway call message.action ...` succeeds.
## Observed behavior
On a production gateway with Telegram accounts running and connected:
- `openclaw gateway call channels.status --... | closed | completed | false | 1 | [] | [] | 2026-05-01T06:36:43Z | 2026-05-03T22:44:21Z | 2026-05-03T22:44:21Z | CONTRIBUTOR | null | 20260504T000036Z | 2026-05-04T00:00:36Z | glfruit | 3,939,154 | MDQ6VXNlcjM5MzkxNTQ= | User | false |
openclaw/openclaw | 4,363,246,759 | I_kwDOQb6kR88AAAABBBHcpw | 75,487 | https://github.com/openclaw/openclaw/issues/75487 | https://api.github.com/repos/openclaw/openclaw/issues/75487 | Heartbeat activeHours timezone ignored — phase schedules based on UTC, not configured timezone | ### Heartbeat activeHours ignores configured timezone — schedules based on UTC phase instead
**Heartbeat 配置了 activeHours 时区为 Asia/Shanghai，但调度逻辑完全按 UTC phase 对齐，activeHours 的时区和窗口被忽略。**
## 配置
```json
"heartbeat": {
"every": "4h",
"activeHours": {
"start": "08:00",
"end": "23:00",
"timezone": "Asia/Sh... | closed | completed | true | 1 | [] | [] | 2026-05-01T06:50:55Z | 2026-05-05T04:26:24Z | 2026-05-02T20:17:55Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | raydoomed | 57,515,812 | MDQ6VXNlcjU3NTE1ODEy | User | false |
openclaw/openclaw | 4,363,293,115 | I_kwDOQb6kR88AAAABBBKRuw | 75,491 | https://github.com/openclaw/openclaw/issues/75491 | https://api.github.com/repos/openclaw/openclaw/issues/75491 | Bug: replaceManagedMarkdownBlock produces duplicate dreaming blocks (regex without g flag) | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
replaceManagedMarkdownBlock uses a regex without the g flag, causing dreaming blocks (e.g. <!-- openclaw:dreaming:light:start -->) to be duplicated in daily memory files when the regex fails to match the existing block (e.g.... | closed | completed | true | 3 | [
"bug",
"regression"
] | [] | 2026-05-01T07:02:51Z | 2026-05-05T04:27:44Z | 2026-05-02T10:28:29Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | asaenokkostya-coder | 272,629,498 | U_kgDOED_--g | User | false |
openclaw/openclaw | 4,363,324,885 | I_kwDOQb6kR88AAAABBBMN1Q | 75,498 | https://github.com/openclaw/openclaw/issues/75498 | https://api.github.com/repos/openclaw/openclaw/issues/75498 | [Bug]: macOS 26.4.1 / 2026.4.29 Telegram Web UI-only replies with partial streaming, polling stall, and session modelOverride pollution | ## Bug type
Regression / delivery-state bug on macOS (not Windows-only)
## Summary
On OpenClaw `2026.4.29 (a448042)` running on macOS, Telegram can enter a polling/transport stall while agent runs continue to complete and write assistant final replies to the local session transcript. The Web UI shows the replies, bu... | open | null | false | 1 | [] | [] | 2026-05-01T07:14:12Z | 2026-05-03T11:50:39Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | sd1114820 | 50,748,585 | MDQ6VXNlcjUwNzQ4NTg1 | User | false |
openclaw/openclaw | 4,363,343,091 | I_kwDOQb6kR88AAAABBBNU8w | 75,501 | https://github.com/openclaw/openclaw/issues/75501 | https://api.github.com/repos/openclaw/openclaw/issues/75501 | [Bug]: too many open files | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
**Bug Report: Skills Watcher causes EMFILE (too many open files) error after recent update**
After upgrading to the latest version of OpenClaw, the Skills Watcher crashes immediately on startup with the following error:
``... | open | null | false | 3 | [
"bug",
"regression"
] | [] | 2026-05-01T07:19:42Z | 2026-05-03T11:50:41Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | wzq-xzwj | 34,700,058 | MDQ6VXNlcjM0NzAwMDU4 | User | false |
openclaw/openclaw | 4,363,428,779 | I_kwDOQb6kR88AAAABBBSjqw | 75,513 | https://github.com/openclaw/openclaw/issues/75513 | https://api.github.com/repos/openclaw/openclaw/issues/75513 | ARM64 performance: redundant loadOpenClawPlugins calls on every request via web provider and capability provider resolution | ## Follow-up to #73075 / PR #73076
After updating to 2026.4.27 (which includes the fixes from PR #73076), response times on ARM64 (Raspberry Pi 4, Node 22) are still severely degraded. Two new independent hot paths call \`loadOpenClawPlugins\` on every request.
## Measured impact (ARM64, Raspberry Pi 4)
| Metric | 2... | closed | completed | true | 2 | [] | [] | 2026-05-01T07:47:23Z | 2026-05-05T04:28:30Z | 2026-05-02T06:40:41Z | CONTRIBUTOR | null | 20260505T060034Z | 2026-05-05T06:00:34Z | jochen | 113,530 | MDQ6VXNlcjExMzUzMA== | User | false |
openclaw/openclaw | 4,363,476,896 | I_kwDOQb6kR88AAAABBBVfoA | 75,522 | https://github.com/openclaw/openclaw/issues/75522 | https://api.github.com/repos/openclaw/openclaw/issues/75522 | [Bug]: | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
## 问题描述
OpenClaw 的启动时间和恢复时间明显偏慢，影响日常使用体验。
### 具体表现
1. **Gateway 启动慢**:
- 执行 `openclaw gateway start` 后，需要多次尝试才能成功
- 经常遇到 "stale gateway process" 需要清理后重试
- 从执行命令到实际可用，耗时可达 30-60 秒
2. **状态探测慢**:
- `openclaw gat... | closed | completed | true | 3 | [
"bug",
"regression"
] | [] | 2026-05-01T08:02:08Z | 2026-05-04T04:35:10Z | 2026-05-01T22:26:22Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | walcherroman749-source | 280,899,652 | U_kgDOEL4wRA | User | false |
openclaw/openclaw | 4,364,554,910 | I_kwDOQb6kR88AAAABBCXSng | 75,654 | https://github.com/openclaw/openclaw/issues/75654 | https://api.github.com/repos/openclaw/openclaw/issues/75654 | [Bug]: Agent prep stages take ~11s per turn, 5x slower than 2026.4.15 | ### Bug type
Crash (process/app exits or hangs)
### Beta release blocker
No
### Summary
After upgrading from 2026.4.15 to 2026.4.29, every agent turn spends ~11s in local prep stages (core-plugin-tools
4.5s, system-prompt 2.7s, stream-setup 2.7s) before any API call. The same setup on 2026.4.15 completes end-to-en... | closed | not_planned | true | 1 | [
"bug",
"bug:crash"
] | [] | 2026-05-01T13:16:49Z | 2026-05-04T04:36:20Z | 2026-05-01T13:21:54Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | A-soldier | 249,521,231 | U_kgDODt9kTw | User | false |
openclaw/openclaw | 4,363,349,368 | I_kwDOQb6kR88AAAABBBNteA | 75,502 | https://github.com/openclaw/openclaw/issues/75502 | https://api.github.com/repos/openclaw/openclaw/issues/75502 | Downgrading from 2026.4.29 to 2026.4.27 fails due to stale file-transfer entry in ~/.openclaw/plugins/installs.json | ## Summary
After upgrading OpenClaw from `2026.4.27` to `2026.4.29`, message handling became extremely slow. I attempted to downgrade back to `2026.4.27` using npm. The downgrade appeared to succeed, but the gateway and CLI could no longer start because OpenClaw kept trying to load a `file-transfer` bundled plugin tha... | open | null | false | 5 | [] | [] | 2026-05-01T07:21:14Z | 2026-05-03T21:43:21Z | null | NONE | null | 20260504T000036Z | 2026-05-04T00:00:36Z | lysopen | 75,906,119 | MDQ6VXNlcjc1OTA2MTE5 | User | false |
openclaw/openclaw | 4,364,571,913 | I_kwDOQb6kR88AAAABBCYVCQ | 75,656 | https://github.com/openclaw/openclaw/issues/75656 | https://api.github.com/repos/openclaw/openclaw/issues/75656 | fix: synchronous session transcript reads block Gateway event loop (WS handshake timeouts, Telegram unresponsive) | ## Summary
The OpenClaw Gateway event loop is severely blocked by synchronous `fs.readFileSync()` calls during agent session preparation. This manifests as WebSocket preauth handshake timeouts (same symptom as #74135, different root cause) and Telegram becoming completely unresponsive for minutes at a time.
The #7413... | closed | completed | true | 4 | [] | [] | 2026-05-01T13:20:05Z | 2026-05-04T04:34:16Z | 2026-05-02T03:19:08Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | DerFlash | 185,013 | MDQ6VXNlcjE4NTAxMw== | User | false |
openclaw/openclaw | 4,364,575,448 | I_kwDOQb6kR88AAAABBCYi2A | 75,658 | https://github.com/openclaw/openclaw/issues/75658 | https://api.github.com/repos/openclaw/openclaw/issues/75658 | feat: sessions cleanup should remove physical .deleted and .reset files, not only index entries | ## Summary
`openclaw sessions cleanup` only manages entries in `sessions.json` (the index), but does not remove the physical `.deleted.*` and `.reset.*` files from disk. These files accumulate indefinitely, causing the session store to grow to hundreds of MB, which worsens the event-loop-blocking issue from synchronou... | open | null | false | 1 | [] | [] | 2026-05-01T13:20:52Z | 2026-05-03T09:13:45Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | DerFlash | 185,013 | MDQ6VXNlcjE4NTAxMw== | User | false |
openclaw/openclaw | 4,364,573,815 | I_kwDOQb6kR88AAAABBCYcdw | 75,657 | https://github.com/openclaw/openclaw/issues/75657 | https://api.github.com/repos/openclaw/openclaw/issues/75657 | fix: local GGUF embedding model warmup blocks Node.js event loop for minutes on startup (ARM64/Pi) | ## Summary
When `memorySearch.provider` is set to `"local"` (the default), the Gateway initializes a GGUF embedding model via node-llama-cpp on the main Node.js thread during startup. On ARM64 hardware (Raspberry Pi), loading a 314 MB GGUF model blocks the event loop for **approximately 6 minutes**, during which the G... | open | null | false | 1 | [] | [] | 2026-05-01T13:20:31Z | 2026-05-04T01:47:49Z | null | NONE | null | 20260504T030031Z | 2026-05-04T03:00:31Z | DerFlash | 185,013 | MDQ6VXNlcjE4NTAxMw== | User | false |
openclaw/openclaw | 4,359,405,683 | I_kwDOQb6kR88AAAABA9dAcw | 75,137 | https://github.com/openclaw/openclaw/issues/75137 | https://api.github.com/repos/openclaw/openclaw/issues/75137 | TUI process consumes 89-99% CPU at idle and during message turns (busy-loop) | # TUI process consumes 89-99% CPU at idle and during message turns (busy-loop)
## Summary
`openclaw-tui` (the local TUI client started via `openclaw chat` / `openclaw tui --local`) sits at **89-99 % CPU** for as long as the window is open — both while idle waiting for user input and while a backend turn is in flight.... | closed | completed | true | 8 | [] | [] | 2026-04-30T15:44:48Z | 2026-05-05T04:26:27Z | 2026-05-02T19:51:35Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | CalyieX | 262,086,326 | U_kgDOD58etg | User | false |
openclaw/openclaw | 4,363,466,995 | I_kwDOQb6kR88AAAABBBU48w | 75,520 | https://github.com/openclaw/openclaw/issues/75520 | https://api.github.com/repos/openclaw/openclaw/issues/75520 | [P0] scoped/unscoped cache-key mismatch causes full plugin reload on every embedded message (20-40s per turn) | ## Summary
`core-plugin-tools` takes 20-32s on every embedded message run because the plugin registry is reloaded from scratch — including re-staging bundled runtime deps (e.g. `acpx` with 31 npm specs). Total prep stages reach 40-70s per message.
**Severity: P0 — completely blocks multi-plugin deployments.** Any gat... | closed | completed | true | 3 | [] | [] | 2026-05-01T07:59:14Z | 2026-05-04T04:35:41Z | 2026-05-01T20:13:52Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | whtoo | 909,131 | MDQ6VXNlcjkwOTEzMQ== | User | false |
openclaw/openclaw | 4,364,514,593 | I_kwDOQb6kR88AAAABBCU1IQ | 75,647 | https://github.com/openclaw/openclaw/issues/75647 | https://api.github.com/repos/openclaw/openclaw/issues/75647 | [bug] regression: discord groupPolicy "allowlist" with no guild.channels block now denies all channels (previously allowed via user filter) | ## Summary
A Discord guild config with `groupPolicy: "allowlist"` and **no `channels` field
under the guild block** previously allowed the bot to operate in any channel
of that guild (filtering only by `users` and `requireMention`). On 2026.4.27
the same config silently denies **every** channel in that guild — the bot
... | closed | completed | true | 1 | [] | [] | 2026-05-01T13:09:08Z | 2026-05-04T04:36:18Z | 2026-05-01T13:38:59Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | jrsdhr | 4,820,276 | MDQ6VXNlcjQ4MjAyNzY= | User | false |
openclaw/openclaw | 4,364,654,879 | I_kwDOQb6kR88AAAABBCdZHw | 75,665 | https://github.com/openclaw/openclaw/issues/75665 | https://api.github.com/repos/openclaw/openclaw/issues/75665 | [Bug]: v2026.4.27 stuck session in one channel blocks all other channels | ## Environment
- Version: v2026.4.27
- Node: 22.22.0
- Provider: Amazon Bedrock
- Channel: Slack socket mode, multiple channels
- Per-channel model override via `modelByChannel`
## Symptom
Long-running Bedrock call in one channel causes shared `sessionId=main` to get stuck (`state=processing`, age 600s+). All other Sl... | open | null | false | 1 | [] | [] | 2026-05-01T13:38:31Z | 2026-05-03T01:23:35Z | null | CONTRIBUTOR | null | 20260503T030030Z | 2026-05-03T03:00:30Z | niceysam | 256,747,835 | U_kgDOD02pOw | User | false |
openclaw/openclaw | 4,364,644,131 | I_kwDOQb6kR88AAAABBCcvIw | 75,664 | https://github.com/openclaw/openclaw/issues/75664 | https://api.github.com/repos/openclaw/openclaw/issues/75664 | [Bug]: memory-wiki CLI status/doctor hang even though direct underlying functions succeed | ## Summary
On OpenClaw 2026.4.29, bundled `memory-wiki` loads and `openclaw wiki --help` works, but `openclaw wiki status` and `openclaw wiki doctor` hang when invoked through the top-level CLI.
The underlying wiki functions complete successfully when called directly from Node in isolated mode, so this appears to be ... | open | null | false | 1 | [] | [] | 2026-05-01T13:36:18Z | 2026-05-03T08:40:31Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | jimmielightner | 8,285,016 | MDQ6VXNlcjgyODUwMTY= | User | false |
openclaw/openclaw | 4,364,670,516 | I_kwDOQb6kR88AAAABBCeWNA | 75,668 | https://github.com/openclaw/openclaw/issues/75668 | https://api.github.com/repos/openclaw/openclaw/issues/75668 | TUI startup extremely slow: model-resolution 11-16s per message + event loop blocked for 5-16 minutes | Hi, I'm experiencing very slow TUI startup with `xiaomi/mimo-v2.5-pro`. Two issues:
## Issue 1: model-resolution takes 11-16s on every message
```
[trace:embedded-run] startup stages:
workspace: 0ms
runtime-plugins: 0-1ms
hooks: 0ms
model-resolution: 11010ms ← always 11-16 seconds
auth: 534ms
attempt-disp... | closed | not_planned | true | 2 | [] | [] | 2026-05-01T13:42:12Z | 2026-05-05T04:28:33Z | 2026-05-02T06:18:43Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | syhdmwi | 227,581,268 | U_kgDODZCdVA | User | false |
openclaw/openclaw | 4,364,693,778 | I_kwDOQb6kR88AAAABBCfxEg | 75,670 | https://github.com/openclaw/openclaw/issues/75670 | https://api.github.com/repos/openclaw/openclaw/issues/75670 | [Bug]: Matrix thread session key case-normalizes event IDs, causing duplicate stuck sessions and thread delivery failures | ### Bug type
Crash (process/app exits or hangs)
### Beta release blocker
No
### Summary
OpenClaw lowercases Matrix event IDs when constructing thread session keys (`sessionKey = ...thread:$<lowercased_event_id>`), but also creates a second session with the original mixed-case event ID. This causes **two compoundin... | open | null | false | 2 | [] | [] | 2026-05-01T13:47:41Z | 2026-05-03T06:51:00Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | jarvis-mns1 | 267,498,334 | U_kgDOD_GzXg | User | false |
openclaw/openclaw | 4,364,660,712 | I_kwDOQb6kR88AAAABBCdv6A | 75,667 | https://github.com/openclaw/openclaw/issues/75667 | https://api.github.com/repos/openclaw/openclaw/issues/75667 | [Feature]: 疑似真实目的被揭穿 | ### Summary
https://chat.deepseek.com/share/kcusrgswyptaedpph5
### Problem to solve
https://chat.deepseek.com/share/kcusrgswyptaedpph5
### Proposed solution
https://chat.deepseek.com/share/kcusrgswyptaedpph5
### Alternatives considered
_No response_
### Impact
https://chat.deepseek.com/share/kcusrgswyptaedpph5... | closed | not_planned | true | 1 | [
"enhancement"
] | [] | 2026-05-01T13:39:52Z | 2026-05-04T04:36:18Z | 2026-05-01T13:54:43Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | shishengh | 168,242,976 | U_kgDOCgcvIA | User | false |
openclaw/openclaw | 4,364,753,441 | I_kwDOQb6kR88AAAABBCjaIQ | 75,673 | https://github.com/openclaw/openclaw/issues/75673 | https://api.github.com/repos/openclaw/openclaw/issues/75673 | qqbot plugin cache causes EACCES permission errors on every gateway restart (macOS) | ## Environment
- macOS 12.7.6 (x64)
- OpenClaw 2026.4.26
- Node 22.22.0
- Installed via npm global
## Problem
On every gateway restart, the following error appears and causes the gateway to hang or fail to become reachable:
```
[channels] failed to load bundled channel qqbot: EACCES: permission denied, unlink '/Users... | closed | completed | true | 1 | [] | [] | 2026-05-01T14:01:35Z | 2026-05-04T04:36:17Z | 2026-05-01T14:06:10Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | MiloGoodBoy | 259,481,780 | U_kgDOD3dgtA | User | false |
openclaw/openclaw | 4,363,375,014 | I_kwDOQb6kR88AAAABBBPRpg | 75,506 | https://github.com/openclaw/openclaw/issues/75506 | https://api.github.com/repos/openclaw/openclaw/issues/75506 | CLI commands warn Cannot find module 'grammy' despite staged Telegram runtime deps | ## Summary
After updating an npm/pnpm OpenClaw install to `2026.4.29`, several CLI commands complete successfully but print repeated warnings like:
```text
[channels] failed to load bundled channel telegram: Cannot find module 'grammy'
Require stack:
- <openclaw package root>/dist/extensions/telegram/send-*.js
```
T... | closed | completed | true | 4 | [] | [] | 2026-05-01T07:29:37Z | 2026-05-04T04:35:38Z | 2026-05-01T20:42:16Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | jarvisz8 | 260,156,352 | U_kgDOD4GrwA | User | false |
openclaw/openclaw | 4,364,258,354 | I_kwDOQb6kR88AAAABBCFMMg | 75,623 | https://github.com/openclaw/openclaw/issues/75623 | https://api.github.com/repos/openclaw/openclaw/issues/75623 | [Bug]: Slack/Telegram providers fail to resolve json5 from plugin-runtime-deps despite json5 being installed | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
OpenClaw 2026.4.29 (a448042) on macOS, Node 25.8.1.
Slack and Telegram providers crash immediately after gateway startup with:
Cannot find package 'json5' imported from /Users/gizmo/.openclaw/plugin-runtime-deps/openclaw-2... | closed | completed | true | 6 | [
"bug",
"regression"
] | [] | 2026-05-01T11:58:47Z | 2026-05-04T04:33:43Z | 2026-05-01T20:42:37Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | cjbubblz | 150,525,580 | U_kgDOCPjWjA | User | false |
openclaw/openclaw | 4,364,773,251 | I_kwDOQb6kR88AAAABBCkngw | 75,676 | https://github.com/openclaw/openclaw/issues/75676 | https://api.github.com/repos/openclaw/openclaw/issues/75676 | High per-turn embedded-run startup/prep overhead on WSL2 even with OpenAI-only plugins | ## Summary
OpenClaw Gateway on WSL2 shows very high per-turn embedded-run startup/prep overhead before the model starts streaming, even with a fresh/minimal setup and only the bundled OpenAI provider enabled.
Tiny webchat messages regularly take ~35-45s before response streaming begins. During that time the gateway N... | closed | duplicate | true | 2 | [] | [] | 2026-05-01T14:06:10Z | 2026-05-04T04:35:06Z | 2026-05-01T22:31:39Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | agaktr | 45,101,051 | MDQ6VXNlcjQ1MTAxMDUx | User | false |
openclaw/openclaw | 4,364,786,857 | I_kwDOQb6kR88AAAABBClcqQ | 75,680 | https://github.com/openclaw/openclaw/issues/75680 | https://api.github.com/repos/openclaw/openclaw/issues/75680 | memory-lancedb autoCapture: MEMORY_TRIGGERS lacks CJK/i18n support | ## Problem
The `autoCapture` feature in `memory-lancedb` plugin uses a hardcoded `MEMORY_TRIGGERS` regex list that only matches English and Czech patterns:
```js
const MEMORY_TRIGGERS = [
/zapamatuj si|pamatuj|remember/i,
/preferuji|radši|nechci|prefer/i,
/rozhodli jsme|budeme používat/i,
/\+\d{10,}/,
/[\w.... | open | null | false | 1 | [] | [] | 2026-05-01T14:09:24Z | 2026-05-03T04:08:36Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | guokewuming | 161,859,935 | U_kgDOCaXJXw | User | false |
openclaw/openclaw | 4,359,165,050 | I_kwDOQb6kR88AAAABA9OUeg | 75,122 | https://github.com/openclaw/openclaw/issues/75122 | https://api.github.com/repos/openclaw/openclaw/issues/75122 | WhatsApp 408 disconnects in 2026.4.27 are caused by event-loop blocking up to 100s, not Baileys | ### Summary
The chronic WhatsApp `status 408 — Connection Terminated` disconnects we see
in `2026.4.27` (also `2026.4.24`) are not a Baileys bug — the new **liveness
diagnostic** subsystem in `2026.4.27` shows the OpenClaw gateway's Node event
loop is being blocked for tens of seconds at a time. While the loop is block... | open | null | false | 1 | [] | [] | 2026-04-30T15:07:24Z | 2026-05-03T03:38:06Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | rutipo | 963,444 | MDQ6VXNlcjk2MzQ0NA== | User | false |
openclaw/openclaw | 4,359,209,239 | I_kwDOQb6kR88AAAABA9RBFw | 75,124 | https://github.com/openclaw/openclaw/issues/75124 | https://api.github.com/repos/openclaw/openclaw/issues/75124 | [Bug]: `command-dispatch: tool` skill slash commands bypass effective tool policy | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
## Why this matters
This is an important security / policy issue, not a cosmetic one.
OpenClaw documents that tool availability should be constrained by:
- tool profiles
- group / channel policy
- sandbox policy
- subagen... | open | null | false | 2 | [
"bug",
"regression"
] | [] | 2026-04-30T15:14:20Z | 2026-05-03T22:24:08Z | null | NONE | null | 20260504T000036Z | 2026-05-04T00:00:36Z | techforgeworks | 275,759,122 | U_kgDOEG_AEg | User | false |
openclaw/openclaw | 4,359,386,400 | I_kwDOQb6kR88AAAABA9b1IA | 75,134 | https://github.com/openclaw/openclaw/issues/75134 | https://api.github.com/repos/openclaw/openclaw/issues/75134 | [Telegram] Heartbeat poll prompt text visible to user in DM conversation | ## Description
The raw `[OpenClaw heartbeat poll]` trigger text is showing up as visible messages in Telegram DM conversations. Users should never see the heartbeat poll prompt — only the agent's response (if any) should be delivered, and `HEARTBEAT_OK` acks should be suppressed by default.
## Expected behavior
Hear... | open | null | false | 1 | [] | [] | 2026-04-30T15:41:54Z | 2026-05-03T06:08:13Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | kteare | 26,414 | MDQ6VXNlcjI2NDE0 | User | false |
openclaw/openclaw | 4,359,697,245 | I_kwDOQb6kR88AAAABA9uzXQ | 75,163 | https://github.com/openclaw/openclaw/issues/75163 | https://api.github.com/repos/openclaw/openclaw/issues/75163 | Bug: TUI mid-session model switch passes raw alias instead of resolved model ID | :label: **Labels to add:** bug, tui
# Bug: TUI mid-session model switch passes raw alias instead of resolved model ID
## Summary
When switching models in the OpenClaw TUI (web dashboard), the model selector writes the display alias (e.g. `"MoE"`, `"Dense"`) directly into the session config without resolving it to a ... | open | null | false | 1 | [] | [] | 2026-04-30T16:30:44Z | 2026-05-04T14:14:11Z | null | NONE | null | 20260504T180039Z | 2026-05-04T18:00:39Z | david-r-jones | 275,712,702 | U_kgDOEG8Kvg | User | false |
openclaw/openclaw | 4,359,731,779 | I_kwDOQb6kR88AAAABA9w6Qw | 75,166 | https://github.com/openclaw/openclaw/issues/75166 | https://api.github.com/repos/openclaw/openclaw/issues/75166 | [Bug]: image/video provider inventory leaks into Discord group channels as visible tool output | ### Summary
`image_generate` / `video_generate` provider inventory output can be delivered into Discord group channels as user-visible tool output, leaking internal provider/model/configured/auth-hint information.
This is not an API-key leak, but it exposes operator environment details that should stay internal, espe... | closed | completed | true | 1 | [] | [] | 2026-04-30T16:36:20Z | 2026-05-05T04:26:01Z | 2026-05-02T23:34:46Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | leoge007 | 116,452,300 | U_kgDOBvDrzA | User | false |
openclaw/openclaw | 4,360,080,549 | I_kwDOQb6kR88AAAABA-GMpQ | 75,187 | https://github.com/openclaw/openclaw/issues/75187 | https://api.github.com/repos/openclaw/openclaw/issues/75187 | [Bug]: Auto-generated `AGENTS.md` puts load-bearing tool-use rules at the bottom; head-truncation by `bootstrapMaxChars` strips them | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
The `AGENTS.md` template auto-generated by `openclaw doctor --fix` orders content with personality/onboarding guidance at the top and the load-bearing `## Red Lines` + tool-use guidance at the bottom; when a us... | open | null | false | 1 | [
"bug",
"bug:behavior"
] | [] | 2026-04-30T17:36:05Z | 2026-05-03T05:00:34Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | camerono | 1,108,491 | MDQ6VXNlcjExMDg0OTE= | User | false |
openclaw/openclaw | 4,360,094,165 | I_kwDOQb6kR88AAAABA-HB1Q | 75,189 | https://github.com/openclaw/openclaw/issues/75189 | https://api.github.com/repos/openclaw/openclaw/issues/75189 | [Bug]: Default `bootstrapMaxChars=20000` + verbose auto-generated bootstrap content degrades tool dispatch on small/mid models | ### Bug type
Behavior bug (incorrect output/state without crash)
### Beta release blocker
No
### Summary
OpenClaw's default `agents.defaults.bootstrapMaxChars: 20000` combined with the verbose content of the auto-generated workspace bootstrap files (~13 KB of content across AGENTS.md, SOUL.md, TOOLS.md, IDENTITY.m... | open | null | false | 1 | [
"bug"
] | [] | 2026-04-30T17:38:42Z | 2026-05-03T04:44:31Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | camerono | 1,108,491 | MDQ6VXNlcjExMDg0OTE= | User | false |
openclaw/openclaw | 4,360,897,111 | I_kwDOQb6kR88AAAABA-4CVw | 75,244 | https://github.com/openclaw/openclaw/issues/75244 | https://api.github.com/repos/openclaw/openclaw/issues/75244 | [Bug]: Plugin TypeBox tool schemas rejected as 'missing parameters object' by describeMalformedPluginTool — memory tools silently dropped every turn | ## Bug type
Behavior bug (incorrect output/state without crash)
## Summary
Plugin tools that use **TypeBox `Type.Object()` schemas** for their `parameters` field are rejected by OpenClaw's `describeMalformedPluginTool` validator with `"missing parameters object"` on every agent turn. The tools are silently dropped, ... | open | null | false | 1 | [] | [] | 2026-04-30T19:58:55Z | 2026-05-03T04:44:34Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | mikronn2 | 105,823,039 | U_kgDOBk67Pw | User | false |
openclaw/openclaw | 4,360,899,673 | I_kwDOQb6kR88AAAABA-4MWQ | 75,245 | https://github.com/openclaw/openclaw/issues/75245 | https://api.github.com/repos/openclaw/openclaw/issues/75245 | registerHook wrapper drops mutable context changes for agent:bootstrap hooks | ## Summary
`registerHook()` currently wraps plugin hook events by shallow-cloning `event.context` before invoking the plugin handler. That breaks mutable hook semantics for hooks such as `agent:bootstrap`, where core intentionally expects plugins to mutate `event.context.bootstrapFiles` before bootstrap files are conv... | open | null | false | 1 | [] | [] | 2026-04-30T19:59:21Z | 2026-05-03T09:49:58Z | null | CONTRIBUTOR | null | 20260503T120038Z | 2026-05-03T12:00:38Z | 100yenadmin | 239,388,517 | U_kgDODkTHZQ | User | false |
openclaw/openclaw | 4,361,001,506 | I_kwDOQb6kR88AAAABA--aIg | 75,250 | https://github.com/openclaw/openclaw/issues/75250 | https://api.github.com/repos/openclaw/openclaw/issues/75250 | Bug: OpenClaw breaks after Homebrew updates due to mixed Homebrew Node/runtime/plugin cache drift | ## Summary
OpenClaw repeatedly becomes unavailable after `brew upgrade` / OpenClaw updates on macOS when the installed service/CLI is backed by Homebrew-managed Node and globally installed OpenClaw packages.
This happened multiple update cycles in a row on my machine. The failures are not a single bad update; they ap... | open | null | false | 2 | [] | [] | 2026-04-30T20:15:31Z | 2026-05-03T06:08:14Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | drarturryzhov-boop | 241,340,210 | U_kgDODmKPMg | User | false |
openclaw/openclaw | 4,361,019,285 | I_kwDOQb6kR88AAAABA-_flQ | 75,252 | https://github.com/openclaw/openclaw/issues/75252 | https://api.github.com/repos/openclaw/openclaw/issues/75252 | fix(mattermost): draft preview overwrites previous content at every transition (data loss) | ## Summary
The Mattermost streaming draft preview (introduced in #47838 / b38988c) overwrites the visible message at every phase transition, causing previously-shown content to disappear when the agent transitions between thinking, partial replies, and tool calls. Users see content like "Thinking…" → partial reply tex... | open | null | false | 2 | [] | [] | 2026-04-30T20:18:45Z | 2026-05-03T01:23:30Z | null | NONE | null | 20260503T030030Z | 2026-05-03T03:00:30Z | katyaclawd | 259,160,789 | U_kgDOD3J61Q | User | false |
openclaw/openclaw | 4,361,255,882 | I_kwDOQb6kR88AAAABA_N7yg | 75,264 | https://github.com/openclaw/openclaw/issues/75264 | https://api.github.com/repos/openclaw/openclaw/issues/75264 | [Bug]: CLI backend (claude-cli/sonnet-4-6) + Telegram: agent stalls during file Read tool call against a wrapper path; only generic "Something went wrong" reaches Telegram | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
Bot agent on the Claude-CLI backend (claude-cli/claude-sonnet-4-6) stalls for ~3 minutes during a file `Read` tool call against a wrapper path and surfaces only the generic "Something went wrong" message to Telegram, with no... | open | null | false | 1 | [
"bug",
"regression"
] | [] | 2026-04-30T21:05:09Z | 2026-05-03T03:38:10Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | satoyski-droid | 280,802,744 | U_kgDOELy1uA | User | false |
openclaw/openclaw | 4,362,052,724 | I_kwDOQb6kR88AAAABA_-kdA | 75,314 | https://github.com/openclaw/openclaw/issues/75314 | https://api.github.com/repos/openclaw/openclaw/issues/75314 | WebChat fresh session shows only injected "New session started" bubble and skips actual startup greeting | ## Summary
In WebChat / Control UI, a fresh session can surface only the injected confirmation bubble:
> `✅ New session started.`
and **not** trigger the expected first-turn startup greeting from the agent.
From the user's perspective, it looks like the assistant stopped greeting them, or that the first input was p... | open | null | false | 1 | [] | [] | 2026-05-01T00:13:27Z | 2026-05-03T08:26:37Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | cslash-zz | 674,096 | MDQ6VXNlcjY3NDA5Ng== | User | false |
openclaw/openclaw | 4,362,173,422 | I_kwDOQb6kR88AAAABBAF77g | 75,323 | https://github.com/openclaw/openclaw/issues/75323 | https://api.github.com/repos/openclaw/openclaw/issues/75323 | MCP stdio leak fix incomplete for Docker-wrapped MCP servers (parallel spawn case) | ## Summary
MCP server resource leak from parallel spawn (specialist agents) — fix from #65694 may be incomplete for Docker-wrapped stdio MCP servers.
## Environment
- **OpenClaw Version**: 2026.4.21 (also applies to 2026.4.25)
- **Operating System**: Ubuntu 22.04 LTS
- **MCP Config**: Custom MCP servers via `plugins.e... | open | null | false | 1 | [] | [] | 2026-05-01T00:49:34Z | 2026-05-03T09:24:30Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | slideshow-dingo | 258,996,445 | U_kgDOD2_43Q | User | false |
openclaw/openclaw | 4,362,174,273 | I_kwDOQb6kR88AAAABBAF_QQ | 75,324 | https://github.com/openclaw/openclaw/issues/75324 | https://api.github.com/repos/openclaw/openclaw/issues/75324 | [Discord] unresolved SecretRef at runtime despite `secrets audit` showing 0 unresolved | ## Summary
`channels.discord.token` is configured as a SecretRef pointing to a valid `filemain` provider key. `openclaw secrets audit` reports 0 unresolved. But at runtime, Discord channel operations fail with `channels.discord.token: unresolved SecretRef "file:filemain:/DISCORD_BOT_TOKEN"`.
This is a regression/frag... | closed | completed | true | 1 | [] | [] | 2026-05-01T00:49:50Z | 2026-05-04T04:36:02Z | 2026-05-01T15:09:52Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | slideshow-dingo | 258,996,445 | U_kgDOD2_43Q | User | false |
openclaw/openclaw | 4,363,508,434 | I_kwDOQb6kR88AAAABBBXa0g | 75,534 | https://github.com/openclaw/openclaw/issues/75534 | https://api.github.com/repos/openclaw/openclaw/issues/75534 | config: skip writeConfigFile when nextHash === previousHash (no-op short-circuit) | ## Summary
`writeConfigFile` in `src/config/io.ts` computes `nextHash` and `previousHash`, sees they're equal (no-op write), but still proceeds with the atomic rename + audit log + suspicious-reason check + file-watcher trigger. On a deployment that calls `config.patch` frequently with semantically-equivalent payloads... | open | null | false | 1 | [] | [] | 2026-05-01T08:13:43Z | 2026-05-03T06:08:19Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | pksidekyk | 264,911,230 | U_kgDOD8o5fg | User | false |
openclaw/openclaw | 4,363,531,153 | I_kwDOQb6kR88AAAABBBYzkQ | 75,538 | https://github.com/openclaw/openclaw/issues/75538 | https://api.github.com/repos/openclaw/openclaw/issues/75538 | [Bug]: QQ Bot streaming repeatedly falls back to static after 2026.4.29 update, likely related to dist pruning regression | [Bug]: QQ Bot streaming repeatedly falls back to static and connection timeouts after 2026.4.29 update, likely related to dist file pruning regression
### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
After updating to OpenClaw 2026.4.29, the QQ Bot channel exhibits severe... | closed | completed | true | 2 | [] | [] | 2026-05-01T08:22:13Z | 2026-05-04T04:35:01Z | 2026-05-01T22:34:30Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | Lunareith | 262,595,192 | U_kgDOD6bieA | User | false |
openclaw/openclaw | 4,363,535,419 | I_kwDOQb6kR88AAAABBBZEOw | 75,539 | https://github.com/openclaw/openclaw/issues/75539 | https://api.github.com/repos/openclaw/openclaw/issues/75539 | Telegram/QQBot plugins fail to connect on servers without global IPv6 (undici IPv6-first issue) | ## Description
After upgrading to OpenClaw v2026.4.29, both Telegram and QQBot plugins fail to connect on cloud servers that lack global IPv6 reachability. The underlying issue is that undici (the HTTP client bundled with plugin-runtime-deps, v8.1.0) defaults to a happy-eyeballs IPv6-first connection strategy, which b... | open | null | false | 2 | [] | [] | 2026-05-01T08:23:47Z | 2026-05-03T03:38:11Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | nakedsleepwalker | 267,792,147 | U_kgDOD_YvEw | User | false |
openclaw/openclaw | 4,363,559,792 | I_kwDOQb6kR88AAAABBBajcA | 75,541 | https://github.com/openclaw/openclaw/issues/75541 | https://api.github.com/repos/openclaw/openclaw/issues/75541 | [Bug] TypeError at prompt assembly stage when lossless-claw is enabled (reading 'length' on undefined) | # Bug Report: `TypeError: Cannot read properties of undefined (reading 'length')` at prompt assembly stage
**To file:** https://github.com/openclaw/openclaw/issues/new
---
## Description
When `@martian-engineering/lossless-claw` v0.9.2 is enabled, all model calls fail at the prompt assembly stage with `TypeError: C... | open | null | false | 1 | [] | [] | 2026-05-01T08:32:19Z | 2026-05-03T08:26:45Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | tyyim | 15,803,885 | MDQ6VXNlcjE1ODAzODg1 | User | false |
openclaw/openclaw | 4,363,812,649 | I_kwDOQb6kR88AAAABBBp_KQ | 75,569 | https://github.com/openclaw/openclaw/issues/75569 | https://api.github.com/repos/openclaw/openclaw/issues/75569 | Personal WhatsApp dmPolicy default ("pairing") leaks pair codes to every contact | ## Summary
When OpenClaw is connected to a **personal WhatsApp account** (using the user's own phone number rather than a dedicated bot account), the default `channels.whatsapp.dmPolicy: "pairing"` causes the gateway to auto-respond to *every unknown sender* — i.e. every personal contact who is not on `allowFrom` — wi... | open | null | false | 1 | [] | [] | 2026-05-01T09:45:03Z | 2026-05-03T05:00:47Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | loicverbeeck-cyber | 280,918,085 | U_kgDOEL54RQ | User | false |
openclaw/openclaw | 4,363,833,006 | I_kwDOQb6kR88AAAABBBrOrg | 75,574 | https://github.com/openclaw/openclaw/issues/75574 | https://api.github.com/repos/openclaw/openclaw/issues/75574 | [Bug]: Telegram network config (dnsResultOrder, autoSelectFamily) not applied to grammy/undici — requests timeout on IPv6-only resolve | ## Bug type
Behavior bug (incorrect output/state without crash)
## Summary
On hosts where IPv6 is unreachable (`curl -6 api.telegram.org` fails with "Network is unreachable") but IPv4 works, Telegram API calls timeout despite correctly configured `channels.telegram.network` settings. The config is read and acknowled... | open | null | false | 1 | [] | [] | 2026-05-01T09:50:59Z | 2026-05-03T08:40:25Z | null | NONE | null | 20260503T090142Z | 2026-05-03T09:01:42Z | jack-jackhui | 65,952,706 | MDQ6VXNlcjY1OTUyNzA2 | User | false |
openclaw/openclaw | 4,364,642,668 | I_kwDOQb6kR88AAAABBCcpbA | 75,663 | https://github.com/openclaw/openclaw/issues/75663 | https://api.github.com/repos/openclaw/openclaw/issues/75663 | Bug: Subagent completion announcements silently drop on Telegram when parent session is idle between turns | ## Summary
When a subagent spawned via `sessions_spawn` completes its work in the background, the completion announcement (result) is silently dropped and never delivered to the Telegram user. The user only sees the result after sending another message that re-triggers the session.
## Environment
- **OpenClaw versio... | open | null | false | 1 | [] | [] | 2026-05-01T13:36:02Z | 2026-05-03T11:50:46Z | null | NONE | null | 20260503T120038Z | 2026-05-03T12:00:38Z | yozakura-ava | 280,745,628 | U_kgDOELvWnA | User | false |
openclaw/openclaw | 4,364,819,458 | I_kwDOQb6kR88AAAABBCncAg | 75,683 | https://github.com/openclaw/openclaw/issues/75683 | https://api.github.com/repos/openclaw/openclaw/issues/75683 | image_generate outputs should be accessible from sandbox workspace for follow-up workflows | ## Summary
`image_generate` returns a `MEDIA:` path under `/home/<user>/.openclaw/media/tool-image-generation/...`, but sandboxed agents cannot read that file or use it as input for follow-up tools such as `exec`, `basecamp attach`, local image processing, OCR/text verification, or archiving.
For agent workflows, gen... | open | null | false | 1 | [] | [] | 2026-05-01T14:16:57Z | 2026-05-03T22:24:09Z | null | NONE | null | 20260504T000036Z | 2026-05-04T00:00:36Z | aaajiao | 49,885 | MDQ6VXNlcjQ5ODg1 | User | false |
openclaw/openclaw | 4,364,852,732 | I_kwDOQb6kR88AAAABBCpd_A | 75,688 | https://github.com/openclaw/openclaw/issues/75688 | https://api.github.com/repos/openclaw/openclaw/issues/75688 | Gateway 100% CPU from startup, node.list 21-35s latency, 724MB memory (v2026.4.29, macOS ARM64) | ## Summary
The OpenClaw Gateway process consumes **100% CPU from the moment it starts** and never drops. Memory usage is **724MB (peak 900MB)**. The `node.list` RPC call takes **21-35 seconds**, and `sessions.list` takes 2-5 seconds. The Web UI (openclaw-control-ui) polls these endpoints every 2-3 seconds, causing req... | closed | completed | true | 4 | [] | [] | 2026-05-01T14:25:05Z | 2026-05-04T04:35:02Z | 2026-05-01T22:34:28Z | CONTRIBUTOR | null | 20260504T060031Z | 2026-05-04T06:00:31Z | 1yihui | 197,358,991 | U_kgDOC8N1jw | User | false |
openclaw/openclaw | 4,364,875,425 | I_kwDOQb6kR88AAAABBCq2oQ | 75,691 | https://github.com/openclaw/openclaw/issues/75691 | https://api.github.com/repos/openclaw/openclaw/issues/75691 | Feature request: openclaw update should detect when invoked from inside the gateway process tree (PPID-ancestry guard) | ## Summary
When `openclaw update --tag <ver> --yes` is invoked from a process whose parent process tree includes the active gateway PID (e.g., from inside an agent's `exec` subprocess), the package install and doctor migrations complete successfully, but the supervised-mode restart silently fails to take effect. The b... | closed | completed | true | 3 | [
"In Progress"
] | [] | 2026-05-01T14:30:08Z | 2026-05-05T04:25:34Z | 2026-05-03T03:40:37Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | islandpreneur007 | 258,422,617 | U_kgDOD2c3WQ | User | false |
openclaw/openclaw | 4,364,877,591 | I_kwDOQb6kR88AAAABBCq_Fw | 75,692 | https://github.com/openclaw/openclaw/issues/75692 | https://api.github.com/repos/openclaw/openclaw/issues/75692 | Feature request: openclaw update should support system-level systemd units (not just user-level) | ## Summary
`openclaw update` generates a detached restart-script that uses `systemctl --user restart <unit>`. Operators running OpenClaw under a system-level systemd unit (`/etc/systemd/system/openclaw.service`, restarted via `sudo systemctl restart openclaw`) are silently bypassed by this restart path: the script eit... | closed | not_planned | true | 1 | [] | [] | 2026-05-01T14:30:37Z | 2026-05-04T04:36:15Z | 2026-05-01T14:34:37Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | islandpreneur007 | 258,422,617 | U_kgDOD2c3WQ | User | false |
openclaw/openclaw | 4,363,206,470 | I_kwDOQb6kR88AAAABBBE_Rg | 75,480 | https://github.com/openclaw/openclaw/issues/75480 | https://api.github.com/repos/openclaw/openclaw/issues/75480 | Slack adapter goes silently dead after channel-stop timeout (post-#56646) | ## Summary
After a `channel stop exceeded 5000ms after abort; continuing shutdown` log line, the Slack provider stops logging anything for hours and stops dispatching inbound events. The gateway process stays alive, `/health` returns OK, but Slack messages to the bot are silently dropped. The only known recovery is a ... | closed | not_planned | true | 1 | [] | [] | 2026-05-01T06:39:24Z | 2026-05-04T04:36:14Z | 2026-05-01T14:34:40Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | jnikolaidis | 232,676,647 | U_kgDODd5dJw | User | false |
openclaw/openclaw | 4,363,464,347 | I_kwDOQb6kR88AAAABBBUumw | 75,519 | https://github.com/openclaw/openclaw/issues/75519 | https://api.github.com/repos/openclaw/openclaw/issues/75519 | Telegram supervisor leaves channel in stopped, disconnected state after stop()-timeout during polling-stall recovery | # Bug: Telegram channel stays in `stopped, disconnected` after stop()-timeout during polling-stall recovery
**Repo:** https://github.com/openclaw/openclaw/issues
**openclaw version:** 2026.4.26 (be8c246)
**Node:** v25.9.0
**OS:** macOS (darwin 24.6.0)
**Channel:** Telegram (default account, polling mode)
## Summary
... | closed | not_planned | true | 1 | [] | [] | 2026-05-01T07:58:28Z | 2026-05-04T04:36:13Z | 2026-05-01T14:34:45Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | ronakdinesh | 25,378,146 | MDQ6VXNlcjI1Mzc4MTQ2 | User | false |
openclaw/openclaw | 4,364,435,915 | I_kwDOQb6kR88AAAABBCQByw | 75,642 | https://github.com/openclaw/openclaw/issues/75642 | https://api.github.com/repos/openclaw/openclaw/issues/75642 | [Performance] Gateway takes 50+ seconds per request - plugin tools loading on every request | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
Gateway takes 50+ seconds per request for simple messages. The typing indicator appears late, and the actual response is very slow.
### Steps to reproduce
1. Send any message to the Gateway via Feishu
2. Observe the delay... | closed | not_planned | true | 4 | [
"bug"
] | [] | 2026-05-01T12:49:56Z | 2026-05-04T04:36:13Z | 2026-05-01T14:34:52Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | aaronpancn | 95,955,040 | U_kgDOBbgoYA | User | false |
openclaw/openclaw | 4,364,820,209 | I_kwDOQb6kR88AAAABBCne8Q | 75,684 | https://github.com/openclaw/openclaw/issues/75684 | https://api.github.com/repos/openclaw/openclaw/issues/75684 | Request: iOS TestFlight Invite | ## Request: iOS TestFlight Invite
**Apple ID:** zousheng1107@163.com
**Gateway:** Mac mini (via Tailscale)
**Use case:** Pair iPhone 14 Pro Max as iOS node
**Region:** China mainland
Thanks! | closed | not_planned | true | 1 | [] | [] | 2026-05-01T14:17:07Z | 2026-05-04T04:36:12Z | 2026-05-01T14:34:57Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | zousheng1107 | 129,469,560 | U_kgDOB7eMeA | User | false |
openclaw/openclaw | 4,364,868,847 | I_kwDOQb6kR88AAAABBCqc7w | 75,689 | https://github.com/openclaw/openclaw/issues/75689 | https://api.github.com/repos/openclaw/openclaw/issues/75689 | [Bug]: Embedded agent initialization blocks event loop for 33s, causing WebSocket handshake timeouts and 45-85% CPU spike | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
### Summary
Embedded agent initialization takes ~33 seconds of synchronous execution on the Node.js main thread, completely blocking the event loop. This causes WebSocket handshake timeouts, "gateway connect failed" errors,... | closed | not_planned | true | 1 | [
"bug",
"regression"
] | [] | 2026-05-01T14:28:40Z | 2026-05-04T04:36:11Z | 2026-05-01T14:34:58Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | linsensong-hue | 263,350,851 | U_kgDOD7JqQw | User | false |
openclaw/openclaw | 4,364,874,836 | I_kwDOQb6kR88AAAABBCq0VA | 75,690 | https://github.com/openclaw/openclaw/issues/75690 | https://api.github.com/repos/openclaw/openclaw/issues/75690 | Stale suppression: qwen3.6-plus blocked on Coding Plan endpoint despite API support | ## Issue
The Qwen provider plugin (`openclaw.plugin.json`) suppresses `qwen3.6-plus` on the Coding Plan endpoints, but the API actually supports this model.
## Evidence
Suppression entries in `~/.npm-global/lib/node_modules/openclaw/dist/extensions/qwen/openclaw.plugin.json`:
```json
{
"provider": "qwen",
"mode... | closed | not_planned | true | 1 | [] | [] | 2026-05-01T14:30:00Z | 2026-05-04T04:36:11Z | 2026-05-01T14:35:28Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | ollielovethemsnacks | 265,828,550 | U_kgDOD9g4xg | User | false |
openclaw/openclaw | 4,364,901,736 | I_kwDOQb6kR88AAAABBCsdaA | 75,694 | https://github.com/openclaw/openclaw/issues/75694 | https://api.github.com/repos/openclaw/openclaw/issues/75694 | [Bug]: Directly after update from 4.22 to 4.29 agents stop responding. | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
version 4.22 is the last stable working version for me, every other version breasks openclaw for me.
Openclaw automatically updated to 4.29 and it again broke the system... as I did not manually update I noticed it after 30... | closed | not_planned | true | 4 | [
"bug",
"regression"
] | [] | 2026-05-01T14:35:59Z | 2026-05-04T04:35:58Z | 2026-05-01T16:08:45Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | najef1979-code | 233,555,716 | U_kgDODevHBA | User | false |
openclaw/openclaw | 4,364,896,358 | I_kwDOQb6kR88AAAABBCsIZg | 75,693 | https://github.com/openclaw/openclaw/issues/75693 | https://api.github.com/repos/openclaw/openclaw/issues/75693 | Bug: skipWhenBusy: true does not defer heartbeat polls during active webchat sessions | ## 环境
- OpenClaw v2026.4.29
- WSL2 Ubuntu, Node.js v24.14.0
- 两种运行模式均复现：`openclaw gateway run` 和 systemd
## 配置
```json
"heartbeat": {
"every": "30m",
"skipWhenBusy": true
}
```
## 预期行为
活跃 webchat 会话中，心跳应被 `skipWhenBusy` 或主队列 busy 检测自动跳过。
## 实际行为
心跳每 1-5 分钟触发一次，插入正在进行的 webchat 对话，打断工作流。即使配置了 `skipWhenBusy: true` ... | closed | not_planned | true | 1 | [] | [] | 2026-05-01T14:34:43Z | 2026-05-04T04:36:10Z | 2026-05-01T14:38:31Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | WhiteGiverMa | 152,406,589 | U_kgDOCRWKPQ | User | false |
openclaw/openclaw | 4,364,917,617 | I_kwDOQb6kR88AAAABBCtbcQ | 75,696 | https://github.com/openclaw/openclaw/issues/75696 | https://api.github.com/repos/openclaw/openclaw/issues/75696 | Codex Computer Use approval denied via MCP elicitation from Discord/OpenClaw session while direct Codex chat works | ## Summary
Codex Computer Use works from a direct Codex Desktop chat, but fails when invoked through an OpenClaw Discord/Codex harness session with:
```text
Computer Use approval denied via MCP elicitation for app 'com.google.Chrome'
```
The same Mac, same Codex app, and same Computer Use helper can read Chrome/Find... | closed | completed | true | 4 | [] | [] | 2026-05-01T14:39:41Z | 2026-05-05T04:28:25Z | 2026-05-02T06:56:29Z | NONE | null | 20260505T060034Z | 2026-05-05T06:00:34Z | mzkmnk | 114,119,396 | U_kgDOBs1S5A | User | false |
openclaw/openclaw | 4,364,964,929 | I_kwDOQb6kR88AAAABBCwUQQ | 75,697 | https://github.com/openclaw/openclaw/issues/75697 | https://api.github.com/repos/openclaw/openclaw/issues/75697 | [bob-test] permission probe - closing, please ignore | Testing PAT write access. Will close immediately. | closed | not_planned | true | 1 | [] | [] | 2026-05-01T14:50:00Z | 2026-05-04T04:36:09Z | 2026-05-01T14:50:06Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | inxaos | 10,542,202 | MDQ6VXNlcjEwNTQyMjAy | User | false |
openclaw/openclaw | 4,364,524,678 | I_kwDOQb6kR88AAAABBCVchg | 75,650 | https://github.com/openclaw/openclaw/issues/75650 | https://api.github.com/repos/openclaw/openclaw/issues/75650 | [Bug]: 2026.4.29 embedded agent reply latency ~40-47s before stream-ready; packaged channel deps missing | ### Bug type
Regression (worked before, now fails)
### Beta release blocker
No
### Summary
After upgrading to OpenClaw 2026.4.29, Telegram and Discord replies consistently wait ~39-47s before stream-ready while logs show repeated embedded-run prep delays and event-loop/cpu liveness warnings.
### Steps to reproduc... | closed | completed | true | 9 | [
"bug"
] | [] | 2026-05-01T13:11:14Z | 2026-05-04T04:35:11Z | 2026-05-01T22:26:21Z | NONE | null | 20260504T060031Z | 2026-05-04T06:00:31Z | jackiehuangtw | 157,334,485 | U_kgDOCWC71Q | User | false |
openclaw/openclaw | 4,360,127,379 | I_kwDOQb6kR88AAAABA-JDkw | 75,192 | https://github.com/openclaw/openclaw/issues/75192 | https://api.github.com/repos/openclaw/openclaw/issues/75192 | Feature Request: Agent Selector in WebChat UI | # Feature Request: Agent Selector in WebChat UI
## 📋 Summary
Add ability to create new chat sessions with agent selection in WebChat UI.
## 🎯 Problem
**Current behavior:**
- Multiple agents can be configured in `openclaw.json`
- WebChat UI shows agent list in dropdown (agent → agents)
- **BUT:** No way to create ... | open | null | false | 1 | [] | [] | 2026-04-30T17:44:21Z | 2026-05-03T05:00:36Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | alextradesk00 | 254,692,233 | U_kgDODy5LiQ | User | false |
openclaw/openclaw | 4,361,646,568 | I_kwDOQb6kR88AAAABA_lx6A | 75,289 | https://github.com/openclaw/openclaw/issues/75289 | https://api.github.com/repos/openclaw/openclaw/issues/75289 | TUI processes peg 100% CPU each and survive terminal close, eating all RAM | ### Summary
`openclaw-tui` processes spin at 100% CPU per process and do not exit when the controlling terminal closes or when the gateway restarts. On a 16 GB Mac mini running the gateway as a LaunchAgent, this repeatedly causes the machine to run out of RAM and swap-thrash until macOS kills processes or the machine ... | open | null | false | 2 | [] | [] | 2026-04-30T22:29:50Z | 2026-05-03T05:00:42Z | null | NONE | null | 20260503T060030Z | 2026-05-03T06:00:30Z | wolfgang-voss | 264,519,572 | U_kgDOD8Q_lA | User | false |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.